Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +237 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,237 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 2.6527161598205566,
|
| 3 |
+
"best_model_checkpoint": "./results/checkpoint-2000",
|
| 4 |
+
"epoch": 0.9996365838885524,
|
| 5 |
+
"eval_steps": 250,
|
| 6 |
+
"global_step": 2063,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.04845548152634767,
|
| 13 |
+
"grad_norm": 0.2021484375,
|
| 14 |
+
"learning_rate": 0.0019926230341909047,
|
| 15 |
+
"loss": 5.5548,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.09691096305269534,
|
| 20 |
+
"grad_norm": 0.484375,
|
| 21 |
+
"learning_rate": 0.0019623193935821215,
|
| 22 |
+
"loss": 4.0801,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.12113870381586916,
|
| 27 |
+
"eval_loss": 3.651378631591797,
|
| 28 |
+
"eval_runtime": 1225.4443,
|
| 29 |
+
"eval_samples_per_second": 47.902,
|
| 30 |
+
"eval_steps_per_second": 1.497,
|
| 31 |
+
"step": 250
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.145366444579043,
|
| 35 |
+
"grad_norm": 0.6796875,
|
| 36 |
+
"learning_rate": 0.0019092830690827923,
|
| 37 |
+
"loss": 3.6511,
|
| 38 |
+
"step": 300
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.19382192610539067,
|
| 42 |
+
"grad_norm": 0.474609375,
|
| 43 |
+
"learning_rate": 0.0018347669274724923,
|
| 44 |
+
"loss": 3.4945,
|
| 45 |
+
"step": 400
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.24227740763173833,
|
| 49 |
+
"grad_norm": 0.51171875,
|
| 50 |
+
"learning_rate": 0.0017405312490731885,
|
| 51 |
+
"loss": 3.3662,
|
| 52 |
+
"step": 500
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.24227740763173833,
|
| 56 |
+
"eval_loss": 3.2901480197906494,
|
| 57 |
+
"eval_runtime": 1223.1792,
|
| 58 |
+
"eval_samples_per_second": 47.991,
|
| 59 |
+
"eval_steps_per_second": 1.5,
|
| 60 |
+
"step": 500
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.290732889158086,
|
| 64 |
+
"grad_norm": 0.90234375,
|
| 65 |
+
"learning_rate": 0.0016288021449895208,
|
| 66 |
+
"loss": 3.2248,
|
| 67 |
+
"step": 600
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.3391883706844337,
|
| 71 |
+
"grad_norm": 0.5859375,
|
| 72 |
+
"learning_rate": 0.001502218970119089,
|
| 73 |
+
"loss": 3.1488,
|
| 74 |
+
"step": 700
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.3634161114476075,
|
| 78 |
+
"eval_loss": 3.054584503173828,
|
| 79 |
+
"eval_runtime": 1225.6755,
|
| 80 |
+
"eval_samples_per_second": 47.893,
|
| 81 |
+
"eval_steps_per_second": 1.497,
|
| 82 |
+
"step": 750
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.38764385221078135,
|
| 86 |
+
"grad_norm": 0.5546875,
|
| 87 |
+
"learning_rate": 0.001363771974184993,
|
| 88 |
+
"loss": 3.0595,
|
| 89 |
+
"step": 800
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.436099333737129,
|
| 93 |
+
"grad_norm": 0.498046875,
|
| 94 |
+
"learning_rate": 0.001216731663648138,
|
| 95 |
+
"loss": 2.9756,
|
| 96 |
+
"step": 900
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.48455481526347666,
|
| 100 |
+
"grad_norm": 0.423828125,
|
| 101 |
+
"learning_rate": 0.0010645715431691336,
|
| 102 |
+
"loss": 2.9173,
|
| 103 |
+
"step": 1000
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.48455481526347666,
|
| 107 |
+
"eval_loss": 2.87882399559021,
|
| 108 |
+
"eval_runtime": 1223.6,
|
| 109 |
+
"eval_samples_per_second": 47.974,
|
| 110 |
+
"eval_steps_per_second": 1.5,
|
| 111 |
+
"step": 1000
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.5330102967898244,
|
| 115 |
+
"grad_norm": 0.431640625,
|
| 116 |
+
"learning_rate": 0.0009108860616831929,
|
| 117 |
+
"loss": 2.853,
|
| 118 |
+
"step": 1100
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.581465778316172,
|
| 122 |
+
"grad_norm": 0.546875,
|
| 123 |
+
"learning_rate": 0.0007593057014319566,
|
| 124 |
+
"loss": 2.7923,
|
| 125 |
+
"step": 1200
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.6056935190793459,
|
| 129 |
+
"eval_loss": 2.7545697689056396,
|
| 130 |
+
"eval_runtime": 1225.0225,
|
| 131 |
+
"eval_samples_per_second": 47.918,
|
| 132 |
+
"eval_steps_per_second": 1.498,
|
| 133 |
+
"step": 1250
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.6299212598425197,
|
| 137 |
+
"grad_norm": 0.392578125,
|
| 138 |
+
"learning_rate": 0.0006134112157875245,
|
| 139 |
+
"loss": 2.7647,
|
| 140 |
+
"step": 1300
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.6783767413688674,
|
| 144 |
+
"grad_norm": 0.30078125,
|
| 145 |
+
"learning_rate": 0.00047664904181190284,
|
| 146 |
+
"loss": 2.7262,
|
| 147 |
+
"step": 1400
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.726832222895215,
|
| 151 |
+
"grad_norm": 0.337890625,
|
| 152 |
+
"learning_rate": 0.0003522498857445232,
|
| 153 |
+
"loss": 2.692,
|
| 154 |
+
"step": 1500
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.726832222895215,
|
| 158 |
+
"eval_loss": 2.6806631088256836,
|
| 159 |
+
"eval_runtime": 1223.5768,
|
| 160 |
+
"eval_samples_per_second": 47.975,
|
| 161 |
+
"eval_steps_per_second": 1.5,
|
| 162 |
+
"step": 1500
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.7752877044215627,
|
| 166 |
+
"grad_norm": 0.302734375,
|
| 167 |
+
"learning_rate": 0.0002431524046570296,
|
| 168 |
+
"loss": 2.6635,
|
| 169 |
+
"step": 1600
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.8237431859479104,
|
| 173 |
+
"grad_norm": 0.33203125,
|
| 174 |
+
"learning_rate": 0.00015193378712876149,
|
| 175 |
+
"loss": 2.6561,
|
| 176 |
+
"step": 1700
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.8479709267110842,
|
| 180 |
+
"eval_loss": 2.655076503753662,
|
| 181 |
+
"eval_runtime": 1225.3417,
|
| 182 |
+
"eval_samples_per_second": 47.906,
|
| 183 |
+
"eval_steps_per_second": 1.498,
|
| 184 |
+
"step": 1750
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.872198667474258,
|
| 188 |
+
"grad_norm": 0.298828125,
|
| 189 |
+
"learning_rate": 8.074887282213439e-05,
|
| 190 |
+
"loss": 2.6481,
|
| 191 |
+
"step": 1800
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.9206541490006057,
|
| 195 |
+
"grad_norm": 0.26953125,
|
| 196 |
+
"learning_rate": 3.127924912435132e-05,
|
| 197 |
+
"loss": 2.6565,
|
| 198 |
+
"step": 1900
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.9691096305269533,
|
| 202 |
+
"grad_norm": 0.416015625,
|
| 203 |
+
"learning_rate": 4.693527335575154e-06,
|
| 204 |
+
"loss": 2.6513,
|
| 205 |
+
"step": 2000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.9691096305269533,
|
| 209 |
+
"eval_loss": 2.6527161598205566,
|
| 210 |
+
"eval_runtime": 1223.4873,
|
| 211 |
+
"eval_samples_per_second": 47.978,
|
| 212 |
+
"eval_steps_per_second": 1.5,
|
| 213 |
+
"step": 2000
|
| 214 |
+
}
|
| 215 |
+
],
|
| 216 |
+
"logging_steps": 100,
|
| 217 |
+
"max_steps": 2063,
|
| 218 |
+
"num_input_tokens_seen": 0,
|
| 219 |
+
"num_train_epochs": 1,
|
| 220 |
+
"save_steps": 250,
|
| 221 |
+
"stateful_callbacks": {
|
| 222 |
+
"TrainerControl": {
|
| 223 |
+
"args": {
|
| 224 |
+
"should_epoch_stop": false,
|
| 225 |
+
"should_evaluate": false,
|
| 226 |
+
"should_log": false,
|
| 227 |
+
"should_save": true,
|
| 228 |
+
"should_training_stop": true
|
| 229 |
+
},
|
| 230 |
+
"attributes": {}
|
| 231 |
+
}
|
| 232 |
+
},
|
| 233 |
+
"total_flos": 4.825965054971085e+17,
|
| 234 |
+
"train_batch_size": 4,
|
| 235 |
+
"trial_name": null,
|
| 236 |
+
"trial_params": null
|
| 237 |
+
}
|